Traditional data historians are placing today's manufacturers at a distinct competitive disadvantage. These legacy systems, built for single plants, fragment operational data across different facilities, equipment, and proprietary platforms. As manufacturers deploy more sensors, production lines, and connected assets, the number of independent data streams they need to manage is exploding. Historians built around fixed tag counts were never designed to scale with this complexity, and that's where blind spots emerge.

"These legacy tools have become bottlenecks because the shape of industrial data and how we process it has changed. Ingestion volumes are higher, cross-system contextual correlation is key, and interfaces remain closed to the tools that need the data," explains Benjamin Corbett, Solutions Engineer at InfluxData. "Manufacturers end up unable to consolidate operational data across plants, integrate it with analytics and AI, or extend it to new use cases."

These limitations manifest in three areas: operational silos, inability to scale with modern telemetry, and closed architectures that make data difficult to access and extend.

1. The silo problem at every plant

Manufacturers are often trapped by fragmented technology stacks. Their data is severely isolated across multiple facilities, different OEM vendors, and independent systems. While a data historian works well within a single plant, manufacturers increasingly need cross-system, cross-facility visibility to support centralized operations, analytics, and AI applications. Without a unified interface, correlating this information is nearly impossible.

"Every plant has its own collection of systems. ERP manages inventory, DCS controls process flows, MES coordinates production, and machine telemetry captures real-time shop floor status. Each system does its job, but they rarely work together. Disconnected data within a single plant quickly becomes an enterprise-level challenge across multiple facilities," says Corbett.

Take tracing a batch number to inventory as an example. This seemingly simple use case might involve generating a specific batch's performance report, querying errors associated with the batch, grouping statistics by manufacturer ID, or viewing incident records for a specific asset type. However, without creating custom integrations for the data historian, such deep analysis would require a large team mired in manual processes.

2. Insufficient scalability

Manufacturers are capturing telemetry data at increasingly finer resolutions, sometimes down to the nanosecond. For most modern consumption patterns, such as building machine learning pipelines, data fidelity is critical. The key is that solutions need to keep up with such high throughput while maintaining full data fidelity.

"The challenge is that modern use cases require an order of magnitude more data than legacy historians," says Corbett. "In the past, sensors or SCADA systems were polled every 30 to 60 seconds. Today we're almost always dealing with contextual signals at least at second-level intervals—by comparison, it's a data tsunami."

Many legacy historians were never designed to handle today's telemetry workloads. Measuring 1000 sensors on 10 machines at sub-second resolution is enough to push most of these systems to their limits. When a historian can't keep up with the influx of telemetry data, the predictive maintenance systems that depend on it also fail. Detecting subtle changes in vibration, temperature, pressure, or current requires complete, high-resolution data, not a platform that is already overwhelmed just trying to ingest data.

Even if manufacturers overcome technical limitations, they often encounter another obstacle: punitive licensing models. Per-tag pricing means every new sensor drives up costs, making scaling increasingly expensive.

"They've invested heavily in data historians, yet they're penalized for every additional industrial signal," Corbett points out. "This is especially painful when equipment changes frequently, but retaining data still holds value."

3. Lock-in effects of closed interfaces

Another complicating factor is that most historians are closed by design. Proprietary formats and vendor-controlled interfaces lock operational data inside the plant network. Whether to support a centralized data science team, connect to a new analytics platform, or empower other business systems, exporting data often becomes a project in itself.

"Because an organization's data science team is typically centralized, their only way to access data is to reach back into the plant network. Every use case becomes an integration project, duplicating effort while raising security and bandwidth concerns, as many remote OT networks are highly egress-only," Corbett explains.

The end result: organizations depend on the system but struggle to scale it easily, and every new analytics initiative becomes just another integration project.

Modernizing beyond the historian

Modernizing a data historian should start with building a business case around operational outcomes, not just a technical replacement. If treated as a one-off infrastructure project, it's likely to stall in pilot purgatory.

The next step is to deploy a platform that uses industry-standard protocols, is purpose-built for time-series data, and can run at the edge while centralizing data. Historian modernization enables manufacturers to leverage their own data to improve operational efficiency and achieve success at scale.

The following questions help highlight the need and value of adopting a time-series platform:

  • How much engineering effort is required to answer questions that span multiple plants or systems?
  • Can your historian keep up with the volume and frequency of telemetry data generated by current operations?
  • How long does it take to make operational data available to analytics or AI teams?

"Historian modernization must become a strategic priority for the organization, not just one person's pet project. When value isn't owned and understood at a high level, processes get compressed, priorities shift, and efforts lose momentum before they bear fruit," Corbett emphasizes. "When done right, the opposite happens—a data foundation built once, on which repeatable, scalable use cases can be layered. In this model, value compounds with scale, rather than costs and effort climbing alongside it."

Learn how a modern time-series platform eliminates these bottlenecks →